guardian-mcp-server/dist/tools/guardian-longread.js

MCP server for The Guardian newspaper

import { LongReadParamsSchema } from '../types/guardian.js';
import { formatArticleResponse, validateDate } from '../utils/formatters.js';
export async function guardianLongread(client, args) {
    const params = LongReadParamsSchema.parse(args);
    // Search for Long Read articles using the specific tag
    const searchParams = {
        tag: 'news/series/the-long-read',
        'page-size': params.page_size || 10,
        page: params.page || 1,
        'show-fields': 'headline,standfirst,body,byline,thumbnail,publication,firstPublicationDate'
    };
    if (params.query) {
        searchParams.q = params.query;
    }
    if (params.from_date) {
        const fromDate = validateDate(params.from_date);
        if (!fromDate) {
            throw new Error(`Invalid from_date format: ${params.from_date}. Use YYYY-MM-DD format.`);
        }
        searchParams['from-date'] = fromDate;
    }
    if (params.to_date) {
        const toDate = validateDate(params.to_date);
        if (!toDate) {
            throw new Error(`Invalid to_date format: ${params.to_date}. Use YYYY-MM-DD format.`);
        }
        searchParams['to-date'] = toDate;
    }
    const response = await client.search(searchParams);
    const articles = response.response.results;
    const pagination = response.response;
    // For search results, default to truncated content for performance
    const formatOptions = { truncate: true, maxLength: 500 };
    return formatArticleResponse(articles, pagination, formatOptions);
}
//# sourceMappingURL=guardian-longread.js.map